A unit status analysis method and system based on network source network-related test

Abstract

The present invention discloses a method and system for analyzing the status of a unit based on a network-source network-related test. The method comprises the following steps: obtaining a network-related test report sent by a power plant; extracting corresponding equipment parameters; inputting the equipment parameters into a correlation calculation formula to calculate the equipment parameter correlation; classifying the equipment parameters according to the equipment parameter correlation calculation result to obtain a set of related parameters; selecting information of a strongly related parameter set from the set of related parameters and calculating the unit's degradation degree value in combination with the test report; determining a weight coefficient and a basic deduction value by the degradation degree value, and then calculating the deduction value of the status quantity to make a final operating state judgment. The present invention can timely update the unit status according to the network-source unit network-related test results, thereby making the network-source unit management more accurate and reasonable, being able to timely identify units with abnormal operating states, and promptly issue early warnings when no failure occurs, urging the implementation of rectification, and ensuring the safety of the power grid.

Description

Technical Field

[0001] The invention relates to a method and system for analyzing a unit state based on a network source network-related test, and belongs to the technical field of power systems. Background Art

[0002] The development of the power grid carries the mission of optimizing the allocation of energy resources, reducing energy consumption, effectively utilizing new energy, and promoting technological progress in emerging industries. It has become an important part of my country's energy strategy. As an important part of the construction of smart grids, with the extensive application of power electronic devices and the commissioning of large-capacity high-voltage direct current transmission, the characteristics of "power electronics" and "strong direct current and weak alternating current" of my country's power grid are becoming increasingly obvious. On the one hand, the new characteristics of the power grid directly affect the operation of the generator set. On the other hand, the capacity of the generator set increases year by year, and its operating conditions also directly affect the safety of the power grid. Under this new situation, it is of great significance to strengthen the coordination research between the generator set and the power grid. The generator grid-related test is a preventive test, an important part of the operation and maintenance of power equipment, and one of the effective means to ensure the safe operation of the power system. It is an important measure to prevent accidents or equipment damage and safe operation.

[0003] The existing technology mainly conducts status assessment on wind turbines, but there is no special evaluation system for traditional turbines, and the data from grid-related tests is not added to the evaluation system for dynamic performance evaluation. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a unit status analysis method and system based on network-source grid-related tests, which can timely update the unit status according to the network-source unit grid-related test results, thereby making the network-source unit management more accurate and reasonable, and can timely identify units with abnormal operating status, and issue early warnings in time when no fault occurs, supervise the implementation of rectification, and ensure the safety of the power grid.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] In a first aspect, the present invention provides a method for analyzing a unit state based on a network source network-related test, comprising:

[0007] Obtain the grid-related test reports sent by the power plant, and convert the test reports in the form of images into editable text information through OCR image recognition technology;

[0008] The converted test report is transferred to the equipment parameter template pre-configured according to different file types to extract the corresponding equipment parameters;

[0009] Input the device parameters into a correlation calculation formula to calculate the device parameter correlation;

[0010] According to the calculation results of the device parameter correlation, the device parameters are classified to obtain a set of associated parameters;

[0011] Selecting information related to the unit operation safety and belonging to the strongly associated parameter set from the associated parameter set and calculating the degradation degree value of the unit in combination with the test report;

[0012] The weight coefficient and the basic deduction value are determined by the degradation degree value, and then the deduction value of the state quantity is calculated to make a final operation state judgment.

[0013] Furthermore, an equipment status assessment model is established to assess the current operating status of the unit, including:

[0014] Establish a state characteristic evaluation model for each unit through historical data of multiple units in multiple periods, that is, a state evaluation model based on Bayesian classifier;

[0015] Through real-time collection of field data, the probability that the current state belongs to different categories is calculated, and the formula is as follows:

[0016] y=f(x)=argmax

[0017]

[0018] The current operating status of the unit is comprehensively evaluated based on the probability that the current status belongs to different categories.

[0019] Furthermore, various indicators of the unit are summarized to form a unit safety assessment report, which is sent to the corresponding system specialists via email.

[0020] Furthermore, the device parameters are input into a correlation calculation formula to calculate the device parameter correlation, including:

[0021] Assume that the two sequences to be calculated for correlation are x and y, respectively, and perform dimensionless processing on the two sequences x and y;

[0022] Find the correlation coefficient ξ(x i ), the formula is as follows:

[0023]

[0024] In the formula, ξ(x i ) represents the serial correlation coefficient, ρ represents the resolution coefficient, Δ(min) represents the minimum difference of the second level, Δ(max) represents the maximum difference of the two levels, Δ 0i (k) represents the absolute difference between each point on the comparison sequence Xi curve and each point on the reference sequence X0 curve;

[0025] Calculate the correlation r between different sequences at a certain timei , the formula is as follows:

[0026]

[0027] In the formula, r i Represents the grey relational degree of sequence x to y, where r i The closer the value is to 1, the better the correlation between sequences x and y.

[0028] Furthermore, the device parameters are classified according to the device parameter association calculation result to obtain the associated parameter set, including:

[0029] According to the calculation results of equipment parameter correlation, a neural network model for classification prediction of parameter status of each component of the unit is constructed;

[0030] Randomly select parameters with low correlation and high correlation as training set and validation set, select a set of parameters with no correlation or low correlation and a set of parameters with strong correlation as test set, input them into the neural network model of classification prediction, and obtain the classification prediction value;

[0031] The classification prediction values ​​are clustered by unsupervised clustering method to obtain the associated parameter set.

[0032] Furthermore, the information related to the unit operation safety and belonging to the strongly associated parameter set is selected from the associated parameter set and combined with the test report to calculate the degradation degree value of the unit, and the formula is as follows:

[0033]

[0034] Where g(x) is the degradation degree of the evaluation index; x is the measured value of the evaluation index; α is the allowable value of the index; and θ is the degradation degree limit.

[0035] Furthermore, the weight coefficient and the basic deduction value are determined by the degradation degree value, and then the deduction value of the state quantity is calculated, and the final operation state score is calculated, including:

[0036] Determine the weight coefficient and the basic deduction value according to the degradation degree value;

[0037] The basic deduction value of the state quantity is multiplied by the weight coefficient to obtain the deduction value of the state quantity;

[0038] Compare the deduction score of the status quantity with the pre-set basic evaluation criteria to make the final operating status judgment.

[0039] In a second aspect, the present invention provides a unit status analysis system based on network source network-related test, comprising:

[0040] The grid-related test report digitization unit is used to obtain the grid-related test reports sent by the power plant and convert the test reports in the form of pictures into editable text information through OCR image recognition technology;

[0041] The test report parameter extraction unit is used to transfer the converted test report to the equipment parameter template pre-configured according to different file types and extract the corresponding equipment parameters;

[0042] A parameter correlation calculation unit, used for inputting the device parameters into a correlation calculation formula to calculate the device parameter correlation;

[0043] The parameter state classification unit is used to classify the device parameters according to the device parameter correlation calculation result and obtain the correlation parameter set;

[0044] A unit for calculating the degradation degree of the unit, used to select information related to the unit operation safety and belonging to the strongly associated parameter set from the associated parameter set and calculate the degradation degree value of the unit in combination with the test report;

[0045] The parameter status evaluation unit is used to determine the weight coefficient and the basic deduction value through the degradation degree value, and then calculate the deduction value of the status quantity to make a final operation status judgment.

[0046] In a third aspect, the present invention provides a unit status analysis device based on a network source network-related test, including a processor and a storage medium;

[0047] The storage medium is used to store instructions;

[0048] The processor is configured to operate according to the instructions to execute the steps of any of the aforementioned methods.

[0049] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the aforementioned methods.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention provides a method and system for analyzing the status of a unit based on a network-source grid-related test, which can timely update the status of the unit according to the network-source unit grid-related test results, thereby making the management of the network-source unit more accurate and reasonable, and can timely identify units with abnormal operating status, and issue early warnings in time when no fault occurs, urge the implementation of rectification, and ensure the safety of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a flow chart of a method for analyzing unit status based on network source and network-related test provided by an embodiment of the present invention;

[0053] Figure 2 It is a schematic diagram of a classification situation element map provided by an embodiment of the present invention;

[0054] Figure 3 It is a screenshot of a test report provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0056] Example 1

[0057] This embodiment introduces a unit status analysis method based on network source network-related test, including:

[0058] Obtain the grid-related test reports sent by the power plant, and convert the test reports in the form of images into editable text information through OCR image recognition technology;

[0059] The converted test report is transferred to the equipment parameter template pre-configured according to different file types to extract the corresponding equipment parameters;

[0060] Input the device parameters into a correlation calculation formula to calculate the device parameter correlation;

[0061] According to the calculation results of the device parameter correlation, the device parameters are classified to obtain a set of associated parameters;

[0062] Selecting information related to the unit operation safety and belonging to the strongly associated parameter set from the associated parameter set and calculating the degradation degree value of the unit in combination with the test report;

[0063] The weight coefficient and the basic deduction value are determined by the degradation degree value, and then the deduction value of the state quantity is calculated to make a final operation state judgment.

[0064] The unit status analysis method based on network source network-related test provided in this embodiment specifically involves the following steps in its application process:

[0065] The present invention includes network source and network-related test result analysis and unit operation status performance evaluation management.

[0066] Analysis of network-related test results: The network-related test results analysis function is achieved by extracting key network-related parameters from the test report, storing them in the database, and analyzing the value of each parameter to determine whether it is within a reasonable range.

[0067] Unit operation status performance evaluation management: Build a unit operation status performance evaluation system, conduct comprehensive statistical analysis on various test reports of the corresponding unit, perform weighted sum calculation on important indicators, derive the unit operation status score, and generate a unit performance evaluation report to be fed back to the system specialist.

[0068] Step 1: Digitization of grid-related test reports: The grid-source coordination management platform processes the grid-related test files uploaded by the power plant and converts the test reports in the form of images into editable text information through OCR image recognition technology.

[0069] Step 2: Test report parameter extraction: Pre-configure the parameter templates to be extracted according to different file types, import the converted test report, and obtain the corresponding parameter information.

[0070] Step 3: Parameter correlation calculation: A device has many parameters, such as a fan, whose parameters may include motor current, motor coil temperature, bearing temperature, vibration value, inlet and outlet medium temperature and flow rate, etc. There are complex correlations between these parameters. By calculating the correlation between parameters, some parameter information that has little effect on the operating status of the device can be eliminated;

[0071] (1) Assume that the two sequences to be associated are x and y.

[0072] (2) Dimensionless processing of the two sequences x and y. Since the two sequences x and y may represent different meanings, their dimensions may not be the same, which makes it inconvenient to compare them, and even impossible to get an accurate conclusion. Therefore, before calculating the correlation, the data needs to be preprocessed, which is usually called dimensionless processing.

[0073] (3) Find the correlation coefficient ξ(x) between the two sequences x and y i ), as shown in formula (3-2). Among them, ρ represents the resolution coefficient, which generally ranges from 0 to 1, and is usually taken as 0.5.

[0074]

[0075] In the formula, ξ(x i ) represents the sequence correlation coefficient, Δ(min) represents the minimum difference of the second level, Δ(max) represents the maximum difference of the two levels, Δ 0i (k) represents the absolute difference between each point on the comparison sequence Xi curve and each point on the reference sequence X0 curve.

[0076] (4) Calculate the correlation r i

[0077] The calculation of this parameter applies formula (3-3), and the calculation result represents the correlation between different sequences at a certain moment.

[0078]

[0079] Where r i ——Grey correlation degree of sequence x to y.

[0080] r i The closer the value is to 1, the better the correlation between sequences x and y.

[0081] Step 4: Parameter status classification: According to the calculation results of the equipment parameter correlation, classify the equipment parameters for status evaluation value calculation. The higher the parameter correlation, the stronger the influence between the two parameters. When the change of one parameter will affect the change of another parameter, a neural network model for parameter status classification prediction of each component of the unit is constructed. Parameters with low correlation and high correlation are randomly selected as training sets and validation sets. A set of parameters with no correlation or low correlation and a set of parameters with strong correlation are selected as test sets, which are input into the neural network for classification prediction to obtain classification prediction values. The classification prediction values ​​are clustered by unsupervised clustering methods to obtain a set of associated parameters. When a parameter in these parameter sets changes, other parameters in the set will also be affected. The unit component status is evaluated on these associated parameter sets.

[0082] Step 5: Calculation of the status evaluation value of each component of the unit: The degradation degree value of the unit is calculated by combining the set information of strongly correlated parameters related to the unit's operation safety and the test report;

[0083]

[0084] Where g(x) is the degradation degree of the evaluation index; x is the measured value of the evaluation index; α is the allowable value (good value) of the index; θ is the degradation degree limit.

[0085] Step 6: Score the status evaluation index of each parameter of the unit equipment. The status quantity is divided into four levels from light to heavy according to the degree of degradation, namely, level Ⅰ, Ⅱ, Ⅲ and Ⅳ, corresponding to the calculated degradation degree values ​​1, 2, 3 and 4, and the corresponding basic deduction values ​​are 2, 4, 8 and 10 points. The deduction value of the status quantity is determined by the degree of degradation and the weight value, that is, the deduction value of the status quantity is equal to the actual deduction value of the status quantity multiplied by the weight coefficient. When the status quantity is normal, 0 points will be deducted, and no points will be deducted if a certain status is not evaluated.

[0086] Table 2 State quantity scoring coefficient table

[0087]

[0088]

[0089] According to the operation status of each component of the unit, the evaluation criteria can be divided into four aspects: basic situation, classification situation, routine test and diagnostic test. According to the above four aspects, the operation status of each component of the unit is judged according to the deduction situation.

[0090] Table 1 Evaluation standards for each component of the unit

[0091]

[0092] Step 7: Establish equipment status assessment model: Clean and reconstruct the operation data of each component of the unit motor, and conduct big data analysis and machine learning modeling on this basis; use Bayesian classifier-based operation to conduct status assessment. The specific steps are as follows:

[0093] (1) State modeling: A state characteristic assessment model can be established for each unit through historical data of multiple units over multiple periods of time, namely, a state assessment model based on the Bayesian classifier.

[0094] (2) Classification probability calculation: Through real-time collection of field data, the probability that the current state belongs to different categories is calculated according to formula (4-1).

[0095]

[0096] (3) Comprehensive status assessment: Comprehensively assess the current operating status of the unit based on the probability that the current status belongs to different categories.

[0097] Step 8: Equipment status information push: Summarize various indicators of the unit to form a unit safety assessment report, and send it to the corresponding system specialist via email.

[0098] This embodiment can timely update the unit status according to the grid-related test results of the grid-source unit, thereby making the management of the grid-source unit more accurate and reasonable, and can timely identify the units with abnormal operating status, issue early warnings in time when no failure occurs, supervise the implementation of rectification, and ensure the safety of the power grid.

[0099] Example 2

[0100] This embodiment introduces a unit status analysis method based on network source network-related test, including:

[0101] (1) Houshi Thermal Power Plant submitted the test reports of the generator of Unit 7 of the power plant: excitation test report, Pss test report, speed regulation test report, phase leading test report, primary frequency regulation test report, AGC test report, AVC test report, such as Figure 3 As shown;

[0102] (2) Test report parameter extraction: Key information parameters were extracted according to different test types. For example, from the PSS test report, the unit status information was extracted, including the excitation equipment model, generator model, excitation mode, etc.; the classification status information included the unit reactive power of 372 Mvar, the unit active power of 600 MW, and the unit rated capacity of 706 MVA.

[0103] (3) Parameter correlation calculation: Based on the correlation calculation results, the parameters are sorted from large to small to obtain a list of key parameters of the unit;

[0104] (4) Parameter status classification: According to the equipment parameter correlation calculation results, the equipment parameters are classified and the status evaluation value is calculated;

[0105] (5) Calculation of the status evaluation value of each component of the unit: Combine the various parameter information of the unit and compare the actual values ​​of each parameter with the data in the evaluation rule base to perform equipment scoring;

[0106]

[0107]

[0108] (6) Scoring of the status evaluation indicators of each parameter of the unit equipment: The degree of degradation of the equipment is obtained based on the scoring results, and the final operating status score is calculated based on the weight ratio.

[0109] (7) Establish equipment status assessment model: assess the current operating status of the unit.

[0110] (8) Operation status report push: Generate a unit operation status health report and send it to the system specialist via email.

[0111] Example 3

[0112] This embodiment provides a unit status analysis system based on network source network-related test, including:

[0113] The grid-related test report digitization unit is used to obtain the grid-related test reports sent by the power plant and convert the test reports in the form of pictures into editable text information through OCR image recognition technology;

[0114] The test report parameter extraction unit is used to transfer the converted test report to the equipment parameter template pre-configured according to different file types and extract the corresponding equipment parameters;

[0115] A parameter correlation calculation unit, used for inputting the device parameters into a correlation calculation formula to calculate the device parameter correlation;

[0116] The parameter state classification unit is used to classify the device parameters according to the device parameter correlation calculation result and obtain the correlation parameter set;

[0117] A unit for calculating the degradation degree of the unit, used to select information related to the unit operation safety and belonging to the strongly associated parameter set from the associated parameter set and calculate the degradation degree value of the unit in combination with the test report;

[0118] The parameter status evaluation unit is used to determine the weight coefficient and the basic deduction value through the degradation degree value, and then calculate the deduction value of the status quantity to make a final operation status judgment.

[0119] Example 4

[0120] This embodiment provides a unit state analysis device based on network source network-related test, including a processor and a storage medium;

[0121] The storage medium is used to store instructions;

[0122] The processor is used to operate according to the instructions to execute the steps of the method according to any one of Embodiment 1.

[0123] Example 5

[0124] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of any method described in Embodiment 1 are implemented.

[0125] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A unit status analysis method based on network source network test, characterized in that: include: Obtain the grid-related test reports sent by the power plant, and convert the test reports in the form of images into editable text information through OCR image recognition technology; The converted test report is transferred to the equipment parameter template pre-configured according to different file types to extract the corresponding equipment parameters; Inputting the device parameters into a correlation calculation formula to calculate the device parameter correlation; According to the calculation results of the device parameter correlation, the device parameters are classified to obtain a set of associated parameters; Selecting information related to the unit operation safety and belonging to the strongly associated parameter set from the associated parameter set and calculating the degradation degree value of the unit in combination with the test report; The weight coefficient and the basic deduction value are determined by the degradation degree value, and then the deduction value of the state quantity is calculated to make a final operation state judgment; The information related to the unit operation safety and belonging to the strongly associated parameter set is selected from the associated parameter set and combined with the test report to calculate the degradation degree value of the unit. The formula is as follows: Where g(x) is the degradation degree of the evaluation index; x is the measured value of the evaluation index; α is the allowable value of the index; and θ is the degradation degree limit.

2. The unit status analysis method based on network source network test according to claim 1 is characterized in that: Establish equipment status assessment model to assess the current operating status of the unit, including: Establish a state characteristic evaluation model for each unit through historical data of multiple units in multiple periods, that is, a state evaluation model based on Bayesian classifier; Through real-time collection of field data, the probability that the current state belongs to different categories is calculated, and the formula is as follows: The current operating status of the unit is comprehensively evaluated based on the probability that the current status belongs to different categories.

3. The unit status analysis method based on network source network test according to claim 1 is characterized in that: Summarize all indicators of the unit to form a unit safety assessment report, which will be sent to the corresponding system specialist via email.

4. The unit status analysis method based on network source network test according to claim 1 is characterized in that: Inputting the device parameters into a correlation calculation formula to calculate the device parameter correlation, including: Set the two sequences for calculating the correlation degree to be x and y respectively, and perform dimensionless processing on the two sequences x and y; Find the correlation coefficient ξ(x i ), the formula is as follows: In the formula, ξ(x i ) represents the serial correlation coefficient, ρ represents the resolution coefficient, Δ(min) represents the minimum difference of the second level, Δ(max) represents the maximum difference of the two levels, Δ 0i (k) represents the absolute difference between each point on the comparison sequence Xi curve and each point on the reference sequence X0 curve; Calculate the correlation r between different sequences at a certain time i , the formula is as follows: In the formula, r i Represents the grey relational degree of sequence x to y, where r i The closer the value is to 1, the better the correlation between sequences x and y.

5. The unit status analysis method based on network source network test according to claim 1 is characterized in that: The device parameters are classified according to the device parameter association calculation results to obtain an associated parameter set, including: According to the calculation results of equipment parameter correlation, a neural network model for classification prediction of parameter status of each component of the unit is constructed; Randomly select parameters with low correlation and high correlation as training set and validation set, select a set of parameters with no correlation or low correlation and a set of parameters with strong correlation as test set, input them into the neural network model of classification prediction, and obtain the classification prediction value; The classification prediction values ​​are clustered by unsupervised clustering method to obtain the associated parameter set.

6. The unit status analysis method based on network source network test according to claim 1 is characterized in that: The weight coefficient and the basic deduction value are determined by the degradation degree value, and then the deduction value of the state quantity is calculated, and the final operation state score is calculated, including: Determine the weight coefficient and the basic deduction value according to the degradation degree value; The basic deduction value of the state quantity is multiplied by the weight coefficient to obtain the deduction value of the state quantity; Compare the deduction score of the status quantity with the pre-set basic evaluation criteria to make the final operating status judgment.

7. A unit status analysis system based on network source network test, characterized in that: include: The grid-related test report digitization unit is used to obtain the grid-related test reports sent by the power plant and convert the test reports in the form of pictures into editable text information through OCR image recognition technology; The test report parameter extraction unit is used to transfer the converted test report to the equipment parameter template pre-configured according to different file types and extract the corresponding equipment parameters; A parameter correlation calculation unit, used for inputting the device parameters into a correlation calculation formula to calculate the device parameter correlation; The parameter state classification unit is used to classify the device parameters according to the device parameter correlation calculation result and obtain the correlation parameter set; The unit degradation degree calculation unit is used to select information related to the unit operation safety and belonging to the strongly associated parameter set from the associated parameter set and calculate the degradation degree value of the unit in combination with the test report. The calculation formula is as follows: In the formula, g(x) is the degradation degree of the evaluation index; x is the measured value of the evaluation index; α is the allowable value of the index; θ is the degradation degree limit; The parameter status evaluation unit is used to determine the weight coefficient and the basic deduction value through the degradation degree value, and then calculate the deduction value of the status quantity to make a final operation status judgment.

8. A unit status analysis device based on network source network-related test, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images